Nicholas Kacher and Stephan Weiler Research Question Broadly Do entrepreneurial projects have tangible benefits in terms of the growth they create Economywide small young firms drive job creation ID: 710810
Download Presentation The PPT/PDF document "Entrepreneurship and Poverty Reduction i..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Entrepreneurship and Poverty Reduction in Rural America
Nicholas Kacher and Stephan WeilerSlide2
Research Question, BroadlyDo entrepreneurial projects have tangible benefits in terms of the growth they create?
Economy-wide, small young firms drive job creation (e.g. Haltiwanger et al. 2017)
Labor productivity growth falls to near zero in firms older than 5 years
(
Alon et al 2017)But Is Growth Inclusive?
2Slide3
Research Question, NarrowlyDo information spillovers from dynamic turnover – openings and
closures of establishments - lead to:Job growth?Poverty reduction?Do results vary across metro/micro/rural areas?
Does
dynamism in
different sectors have different impacts on poverty reduction?
3Slide4
Conceptual Framework
4
Entrepreneurship
Information
Growth
Schumpeter (1942)
Aghion
& Howitt (1990)
Akerlof (1970)
Hendricks and
Kovenock
(1989)
Lang and Nakamura (1993)
?
Bunten
et al. (2015)
Poverty Reduction
?Slide5
EntrepreneurshipMeasured as the opening (or closure) of a business Establishment“A single physical location where business is conducted or industrial operations are performed” with at least one employee
New firm, or expansion of existing firmEntrepreneurial ProjectNew combination of product/service, location, labor,
land, capital
…
5Slide6
Establishment Dynamism
6
Establishment births and deaths closely correlated, but overall
dynamism varies
widely across countiesDynamism effects?Negatives:
Risk and uncertainty + Frictional unemploymentPositive: Information spillovers, leading to long-term growth?
Data: Survey of Business Owners – Business Information Tracking SeriesSlide7
Declining Dynamism
7
Last decade has only accelerated decrease in dynamism
Openings
and closures have fallen in
Metro and Nonmetro areasIntriguingly slower decrease in Nonmetro countiesSlide8
Information Spillovers
Biz success depends crucially on InformationHypothesis
:
Successes
and failures of other ventures inform followersInformation is market-specificSupply (Factors of Prodn) and Demand (Market)Geographic Informational Asymmetries
Thin information markets deter investmentStagnation can become self-reinforcing
8Slide9
Entrepreneurship and Growth
Bunten et al. (2015) Dynamism as defined by Establishment Births*Deaths alongside Births and
Deaths
contributes to stronger county employment growth in 2000-07 time period
Strongest among Metro countiesEndogeneity addressed by IVsPost-Recession – Dynamism continues to contribute to growth across counties, still strongest in Metros
9Slide10
Entrepreneurship and Poverty
Reduction
Does Dynamism
R
educe Local Poverty?Shambaugh et al. 2018 – Dynamism increases number of job openings, which improves workers’ options and bargaining powerLocal dynamism may be especially important as interstate mobility declines (e.g. Frey 2009)Partridge & Rickman 2006
Increasing returns (e.g. Ades and Glaeser 1994, Krugman 2001) vs. Inter-regional equilibration (e.g. Blanchard and Katz 1992)10Slide11
Poverty – Summary Stats
Poverty data from Census Small Area Income and Poverty Estimates (SAIPE), using Census poverty line definitions
11
Modal poverty rate ~12%, some counties much higher
Poverty declined on average from 2010 to 2016, but not smoothlySlide12
Poverty – Summary Stats
12
Higher average poverty, and less poverty reduction,
in rural counties
Large variation in both initial poverty and poverty reduction
No correlation between 2010 poverty and poverty reduction 2010-2016Slide13
Poverty – Summary Stats
13
Highest poverty rates in
Nonmetro
areasNonmetro Mean poverty
>> Median, thus some with very high poverty ratesPoverty reduction has been slowest in more rural areas
(
back
)Slide14
Empirical ModelOLS first-difference approach:
Entrepreneurial Projects
includes Establishment Births, Deaths and
Births*Deaths
,
together representing
Dynamism
14Slide15
Entrepreneurship and Poverty Reduction – All Counties
Dependent Variable: Percent
Reduction
in County Poverty Rate 2010-2016
All Counties
(n=3032)
Metro
(n=825)
Non-metro
(n=2225)
Establishment Births 2010
1.521***
1.770
1.406***
Establishment Deaths 2010
-0.517
-0.662
0.062
Establishment Births*Deaths 2010
-0.059
-0.048
-0.018
Bartik Demand Shock
-0.245***
-0.370***
-0.158***
Lagged Employment Growth, 1990-2000
0.073***
0.076***
0.058***
Log Employment 2010
0.225
0.350
1.757***
Population Density 2010
-0.168***
-0.190***
5.390
Creative Class Employment Share 2007-2011
-0.520***
-0.391**
-0.521***
Arts Employment Share 2007-2011
0.238
0.219
-1.301**
Share with Bachelor's Degree or Higher, 2010
0.720***
0.695***
0.478***
Lagged Population Growth, 1950-1960
-3.672***
-4.109***
3.834***
Median Age 2010
-0.163***
-0.221**
0.002
Amenity Score
0.506***
0.586
-0.451***
Distance to Nearest MSA
-3.229***
-8.871***
-1.491***
Constant
6.096*
3.058***
-13.66***
R-Squared
0.151
0.211
0.079
Establishment Births
reduce poverty overall, yet…
Effects concentrated in Non-metro…
…in fact, in precisely the
most rural non-core counties
.Slide16
Entrepreneurship and Poverty Reduction
Dependent Variable: Percent
Reduction
in County Poverty Rate 2010-2016 (
obs
= 3032)
All Counties
n=3032
Metro
n=825
Micro
n=637
Rural
n=1610
Establishment Births 2010
1.521***
1.770
1.806
0.805*
Establishment Deaths 2010
-0.517
-0.662
-0.044
-0.222
Establishment Births*Deaths 2010
-0.059
-0.048
-0.022
0.000
Bartik Demand Shock
-0.245***
-0.370***
-0.084
-0.373***
Lagged Employment Growth, 1990-2000
0.073***
0.076***
0.143***
0.031***
Log Employment 2010
0.225
0.350
1.545
2.110***
Population Density 2010
-0.168***
-0.190***
4.934
-1.532
Creative Class Employment Share 2007-2011
-0.520***
-0.391**
-0.808***
-0.298**
Arts Employment Share 2007-2011
0.238
0.219
-3.363**
-0.600
Share with Bachelor's Degree or Higher, 2010
0.720***
0.695***
0.646***
0.451***
Lagged Population Growth, 1950-1960
-3.672***
-4.109***
2.872
5.172***
Median Age 2010
-0.163***
-0.221**
0.065
-0.075
Amenity Score
0.506***
0.586
-0.648***
-0.151
Distance to Nearest MSA
-3.229***
-8.871***
-2.632**
-0.212
Constant
6.096*
3.058***
-10.529
-16.541
R-Squared
0.151
0.211
0.079
0.112**Slide17
Sector-Specific Entrepreneurship and Poverty Reduction
Does the effect vary by sector?
Same model as before, but run separately for each sector
j
defined at the two-digit NAICS level
17Slide18
Sector-Specific Dynamism and Poverty ReductionBirths have a
broadly positive impact on poverty reduction, but does Birth & Death dynamism have additional effects?
Positive
: Coefficient on Births*Deaths is positive at 5% level, or coefficient on Deaths is positive at 5% level
Negative: Coefficient on Births*Deaths is negative at 5% level; Deaths either negative or insignificantHigh/low wage sectors, but also…Tradable/Nontradable (Moretti 2010
) multiplier debateVan Dijk (2018) suggests substantial overestimation
18Slide19
Sector-Specific Dynamism and Poverty Reduction – All Counties
19
Positive
Negative
Mining, Quarrying, and Oil and Gas Extraction
($1,787)
Wholesale Trade ($1,257)
Utilities ($1,716)
Retail Trade ($532)
Manufacturing ($1,129)
Transportation and Warehousing ($532)
Finance and Insurance ($1,530)
Information ($1,257)
Real Estate, Rental, and Leasing
($880)
Administrative and Support and Waste Management and Remediation Services ($656)
Management of Companies ($1,926)
Health Care and Social Assistance ($841)
Other Services (Except Public Administration) ($621)
Arts, Entertainment, and Recreation ($576)
Accommodation and Food Services ($345)
Dynamism in
higher-paying sectors tends to correlate with subsequent poverty reductionSlide20
Sector-Specific Dynamism and Poverty Reduction – Non-Metro Counties
20
Positive
Negative
Construction ($985)
Information ($1,257)
Wholesale Trade ($1,257)
Accommodations and Food Services ($345)
Finance and Insurance ($1,530)
Professional, Scientific, and Technical Services ($1,526)
Administrative and Support and Waste Management and Remediation Services ($656)
Educational Services ($862)
Health Care and Social Assistance ($841)
Other Services (except Public Administration) ($621)
Manufacturing ($1,129)
Retail Sales ($532)
Sectoral dynamism effects much broader in
Nonmetro
areasSlide21
Implications and Next Steps
Dynamism influences local employment growth
Establishment
births increase
both employment growth and reduce poverty in rural areasMany sectors’ dynamism including deaths reduce poverty, especially in rural areasNext stepsFurther analysis of
sectoral effects: Tradable/NontradableSpatial robustness: Does info spill across counties?Possible endogeneity: Instrumental variable approaches?21Slide22
Entrepreneurship and Employment Growth
Dependent Variable: Percent
Employment Growth 2007-2014
All Counties
n=3032
Metro
n=825
Micro
n=637
Rural
n=1610
Establishment Births 2007
4.642***
4.699***
5.324***
1.978***
Establishment Deaths 2007
-2.012***
-5.097***
-0.984
-0.172
Establishment Births*Deaths 2007
-0.0868
0.247**
-0.411**
-0.170**
Bartik
Demand Shock
0.0704***
0.088
0.036
0.105**
Lagged Employment Growth, 1990-2000
0.0321***
0.035***
-0.032
-0.020**
Log Employment 2007
-3.399***
0.112
-1.589*
-1.377***
Log Income 2007
3.408***
1.059
-7.857**
-2.053
Population Density 2007
0.122***
0.123***
13.991
-1.666
Creative Class Employment Share 2007
-0.201***
-0.159*
0.007
0.056
Arts Employment Share 2007
-1.285***
-1.232**
0.096
-0.970*
Share with Bachelor's Degree or Higher, 2000
0.406***
0.376***
0.338**
0.432***
Lagged Population Growth, 1950-1960
-0.767***
-0.768*
3.211*
1.493
Median Age 2000
-0.45***
-0.421***
-0.094
-0.482***
Amenity Score
0.077*
0.153**
-0.426**
0.036
Distance to Nearest MSA
0.000558
0.373
3.930***
0.851
Constant
-2.972
2.909
82.191**
44.460***
R-Squared
0.368
0.494
0.135
0.112
(
back
)Slide23
Sector-Specific Entrepreneurship and Poverty Reduction – Metro Counties (n=825)
23
Positive
Negative
Mining, Quarrying, and Oil and Gas Extraction ($1,787)
Construction ($985)
Wholesale Trade ($1,257)
Information ($1,257)
Finance and Insurance ($1,530)
Real Estate, Rental, and Leasing
($880)
Administrative and Support and Waste Management and Remediation Services ($656)
Educational Services ($862)
Health Care and Social Assistance ($841)
Arts, Entertainment, and Recreation ($576)
Accommodation and Food Services ($345)
Other Services (Except Public Administration) ($621)
Manufacturing ($1,129)
Retail Trade ($532)Slide24
Sector-Specific Entrepreneurship and Poverty Reduction – Micropolitan Counties (n=637)
24
Positive
Negative
Wholesale Trade ($1,257)
Information ($1,257)
Other Services (except Public Administration) ($621)Slide25
Sector-Specific Entrepreneurship and Poverty Reduction – Rural Counties (n=1610)
25
Positive
Negative
Construction ($985)
Agriculture, Forestry, Fishing and Hunting ($545)
Wholesale Trade ($1,257)
Information ($1,257)
Finance and Insurance ($1,530)
Arts, Entertainment, and Recreation ($576)
Real Estate, Rental, and Leasing
($880)
Transportation and Warehousing ($532)
Professional, Scientific, and Technical Services ($1,526)
Administrative and Support and Waste Management and Remediation Services ($656)
Educational Services ($862)
Health Care and Social Assistance ($841)
Other Services (Except Public Administration) ($621)
Manufacturing ($1,129)
Retail Trade ($532)Slide26
Entrepreneurship and Inequality?
Does dynamism effect overall inequality?
Types of jobs
created,
in/out-migration, etc.Data issue – Ginis only available in 5-year ACSDynamism appears to increase inequalityEffects concentrated in Metros and Micros
26Slide27
Entrepreneurship and Inequality
Dependent Variable: Gini Coefficient, 2016 ACS 5-year estimate
All Counties
n=3032
Metro
n=825
Micro
n=637
Rural
n=1610
Establishment Births 2007
-0.0012
-0.0013
0.0002
0.0011
Establishment Deaths 2007
-0.0034**
-0.0059*
0.0051**
0.0004
Establishment Births*Deaths 2007
0.0007***
0.0011**
-0.0005
-0.0002
Bartik
Demand Shock
0.0004*
0.0003
0.0013***
0.0005***
Lagged Employment Growth, 1990-2000
-0.0002***
-0.0002***
-0.0004***
-0.0001***
Log Employment 2007
0.0060***
0.0054***
0.0004
-0.0008
Log Income 2007
0.0254***
0.0470***
-0.0551***
-0.0225***
Population Density 2007
0.0007***
0.0003
0.0548**
0.0699***
Creative Class Employment Share 2007
-0.0012***
-0.0012***
0.0008
0.0002
Arts Employment Share 2007
0.0137***
0.0167***
0.0080**
0.0074***
Share with Bachelor's Degree or Higher, 2000
0.0003
0.0000
-0.0007
-0.0007***
Lagged Population Growth, 1950-1960
-0.0092***
-0.0067***
-0.0228***
-0.0192***
Median Age 2000
-0.0002
-0.0001
-0.0017***
-0.0011***
Amenity Score
0.0004**
-0.0001
0.0032***
0.0034***
Distance to Nearest MSA
-0.0048***
-0.0341***
-0.0008
0.0025*
Constant
0.1496***
-0.0637
1.0355***
0.7131***
R-Squared
0.537
0.631
0.199
0.123